1-(2-hydroxy-2-methyl-3-phenoxypropanoyl)indoline-4-carbonitrile derivatives as potent and tissue selective androgen receptor modulators

Article abstract of DOI:10.1021/jm401625b

We present a novel series of selective androgen receptor modulators (SARMs) which shows excellent biological activity and physical properties. 1-(2-Hydroxy-2-methyl-3-phenoxypropanoyl)-indoline-4-carbonitriles showed potent binding to the androgen receptor (AR) and activated AR-mediated transcription in vitro. Representative compounds demonstrated diminished activity in promoting the intramolecular interaction between the AR carboxyl (C) and amino (N) termini. This N/C-termini interaction is a biomarker assay for the undesired androgenic responses in vivo. In orchidectomized rats, daily administration of a lead compound from this series showed anabolic activity by increasing levator ani muscle weight. Importantly, minimal androgenic effects (increased tissue weights) were observed in the prostate and seminal vesicles, along with minimal repression of circulating luteinizing hormone (LH) levels and no change in the lipid and triglyceride levels. This lead compound completed a two week rat toxicology study, and was well tolerated at doses up to 100 mg/kg/day, the highest dose tested, for 14 consecutive days.

Full text of DOI:10.1021/jm401625b

Article
pubs.acs.org/jmc
1‑(2-Hydroxy-2-methyl-3-phenoxypropanoyl)indoline-4-carbonitrile
Derivatives as Potent and Tissue Selective Androgen Receptor
Modulators
Eugene L. Piatnitski Chekler,* Rayomond Unwalla, Taukeer A. Khan, Raghuram S. Tangirala,
Mark Johnson, Michael St. Andre, James T. Anderson, Thomas Kenney, Sue Chiparri, Chris McNally,
Edward Kilbourne, Catherine Thompson, Sunil Nagpal, Gregory Weber, Scott Schelling, Jane Owens,
Carl A. Morris, Dennis Powell, Patrick R. Verhoest, and Adam M. Gilbert
BioTherapeutics Medicinal Chemistry, Pfizer Global Research and Development, 200 Cambridgepark Drive, Cambridge,
Massachusetts 02140, United States
S
* Supporting Information
ABSTRACT: We present a novel series of selective androgen
receptor modulators (SARMs) which shows excellent bio-
logical activity and physical properties. 1-(2-Hydroxy-2-
methyl-3-phenoxypropanoyl)-indoline-4-carbonitriles showed
potent binding to the androgen receptor (AR) and activated
AR-mediated transcription in vitro. Representative compounds
demonstrated diminished activity in promoting the intra-
molecular interaction between the AR carboxyl (C) and amino
(N) termini. This N/C-termini interaction is a biomarker assay
for the undesired androgenic responses in vivo. In orchidectomized rats, daily administration of a lead compound from this series
showed anabolic activity by increasing levator ani muscle weight. Importantly, minimal androgenic effects (increased tissue
weights) were observed in the prostate and seminal vesicles, along with minimal repression of circulating luteinizing hormone
(LH) levels and no change in the lipid and triglyceride levels. This lead compound completed a two week rat toxicology study,
and was well tolerated at doses up to 100 mg/kg/day, the highest dose tested, for 14 consecutive days.
Although ARs are expressed in numerous tissues, very little is
known about how to selectively target small molecules to a
specific tissue (e.g., muscle vs prostate), thus designing a
muscle selective SARM would be a significant challenge.
However, a disclosure by Ghali et al. suggested an alternative
approach to achieve functional selectivity of AR-dependent
signaling. They identified mutations in the AR ligand-binding
domain (LBD) of patients with androgen insensitivity
syndrome (AIS) in which an insensitivity to androgen receptor
agonists impairs maturation of male sexual tissues.7 Those
mutations prevented AR amino/carboxyl N/C-terminal inter-
actions, and the authors developed a mammalian cell-based
two-hybrid assay that could test for androgen receptor
modulators which displayed this profile. Thus, a compound
with potent AR agonism but with minimal AR N/C-interaction
activity might profile with similar properties as an actual
muscle-selective AR modulator in vivo.
INTRODUCTION
■
The 3C subfamily of nuclear hormone receptors is composed of
the glucocorticoid receptor (GR1), the mineralocorticoid
receptor (MR), the progesterone receptor (PR), and the
androgen receptor (AR).1 The AR shares a similar signaling
mechanism as these steroidal nuclear receptors via binding to
DNA and regulating the transcription of responsive genes. AR
signaling is involved in numerous human biological functions
including male reproductive initiation and differentiation, the
regulation of spermatogenesis, the control of male sexual
behavior, and regulation of the development and maintenance
of the musculoskeletal and adipose tissues.2,3 The endogenous
AR ligands, testosterone and dihydrotestosterone (DHT), are
potent activators of AR signaling and, influenced by the
widespread distribution of the AR, affect multiple tissues. The
pharmaceutical industry has been exploring the use of agents
that selectively activate AR signaling in muscle and bone for a
number of years.4 These agents, called selective androgen
receptor modulators (SARMs), have the potential to treat
multiple muscle wasting diseases and conditions, including
sarcopenia, frailty, and cachexia (as a comorbidity in multiple
diseases), if their desirable anabolic effects could be dissociated
from their undesirable androgenic actions (e.g., hypertrophy of
the prostate).5,6
Previous disclosures by others have shown that this screening
strategy can identify SARMs with anabolic effects with minimal
androgenic activity: Merck (1)8 and GTx (2)9 (Figure 1).
Others have disclosed functionally selective SARMs without
disclosing N/C-terminal interaction assay data for their lead
Received: November 12, 2013
Published: February 17, 2014
© 2014 American Chemical Society
2462
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
Figure 1. SARM clinical candidates and leads.
a
Scheme 1. (S)-1-(2-Hydroxy-2-methyl-3-phenoxypropanoyl)indoline-4-carbonitrile Derivatives
a
Reagents and conditions: (a) NaOH, H2O, acetone, 0−25 °C, 6 h; (b) NBS, DMF, CCl4, 0−25 °C, 48 h; (c) 24% HBr, reflux, 2 h; (d) indoline-4-
carbonitrile, SOCl2, DMA, −12 to 25 °C, 16 h; (e) K2CO3, 2-butanone, 25 °C, 6 h; (f) R-PhOH, K2CO3, 2-butanone, reflux, 48 h.
compounds: Radius (3),10 BMS (4),11 Ligand,12 and Galapagos
(5a, 5b).13,14 We used a similar N/C-interaction assay in our
screening strategy to discover SARMs with excellent in vivo
muscle anabolic activity but with minimal androgenic effects.
The following report outlines our efforts to identify and
optimize a series of functionally selective 1-(2-hydroxy-2-
methyl-3-phenoxypropanoyl) indoline-4-carbonitriles 6, which
was identified through a structure-based design effort around
published and internally generated crystal structures of the AR
LBD.
Molecular Modeling. In an attempt to understand the
binding mode of our indoline series, we docked16 compound
15 (Table 1) using the previously published X-ray structure17
of the AR ligand binding domain (PDB code: 2AXA). The
protein was first prepared using the Protein Preparation wizard
and the docking studies were performed using the Glide
software within Maestro16 (see Supporting Information for
details). The best scoring pose of 15 from the docking studies is
shown in Figure 2 along with the critical residues within the
binding site. The nitrile group of the ligand makes hydrogen
bond interactions to Arg752 and Gln711 residues directly and
also by way of a water mediated interaction. This pattern of
hydrogen bond interaction is similar to that seen with the nitrile
group of (R)-bicalutamide. The pyrrolidine ring is oriented
toward the Leu707 residue to pick up additional hydrophobic
interactions. The (S)-OH group is able to pick up hydrogen
bond interaction with the Asn705 residue, while the methyl
group is oriented toward the Thr877 residue. Docking of the
(R)-OH enantiomer shows a similar binding mode, however,
the hydroxyl group in this isomer is not able to make a
hydrogen bond interaction with Asn705 residue. This differ-
ential hydrogen bond interaction between the isomers could
explain the loss of potency for the (R)-isomer. The phenoxy
group is oriented toward the ligand induced pocket and stacked
RESULTS
■
Chemistry. 1-(2-Hydroxy-2-methyl-3-phenoxypropanoyl)-
indoline-4-carbonitrile (4-cyanoindoline) SARMs may be
prepared as shown in Scheme 1. (R)-Proline 7 and
methacryloyl chloride 8 were mixed under basic conditions to
produce methacrylamide 9. Bromolactonization in the presence
of NBS produces bromolactone 10. Acidic hydrolysis with
aqueous HBr then produces (R)-3-bromo-2-hydroxy-2-methyl-
propanoic acid 11.15 Amide coupling with indoline-4-carbon-
itrile using HATU/DIPEA produces indoline amide 12.
Epoxide formation is carried out using K2CO3 to produce 13,
which is converted to the optically pure target compounds 6 by
the reaction with substituted phenols under basic conditions.
2463
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
various AR X-ray structures, the flexibility of the Trp741
residue is critical for accommodation of diverse AR ligands
within the binding pocket. The repositioning of this residue
results in an expansion of the binding pocket for direct ligand
interaction with residues in the AF-2 domain. It is interesting to
note that in AR X-ray structures complexed to steroidal
agonists such as DHT and R1881, the Trp741 residue is in a
position which completely encloses the binding cavity. This
structural aspect is an important point of differentiation
between nonsteroidal and steroidal AR agonists.
Table 1. 1-(2-Hydroxy-2-methyl-3-
phenoxypropanoyl)indoline-4-carbonitrile 6 Chemical Class
a
in Vitro Biological Data
AR
In Vitro Data. Compounds were assayed for in vitro
anabolic activity using an androgen response element-luciferase
(ARE-LUC) assay in CV-1 cells which measures the ability to
induce transcriptional activity. Agonist function in the ARE-
LUC assay predicts in vivo muscle activity. In vitro androgenic
activity was assessed using an N/C-interaction assay in CV-1
cells.7 Binding to the androgen receptor was assayed by
displacement of [3H] mibolerone from COS cells transfected
with the androgen receptor. ARE-LUC and N/C interaction
assay functional data and AR binding data are presented in
Table 1. Substitution of the terminal Ph group of the (S)-
enantiomer 1-(2-hydroxy-2-methyl-3-phenoxypropanoyl)-
indoline-4-carbonitrile was shown to have a significant effect
on both functional and binding data (Table 1). LipE is a
measure of lipophilic ligand efficiency, and it proved to be a
useful parameter to assess true changes in potency that did not
come from hydrophobic properties of a ligand. The higher LipE
value would correspond to a desirable change of potency as a
result of protein−ligand interactions. The parent Ph analogue
14 shows weak ARE-LUC agonism, weak AR binding, and is
devoid of androgenic activity. The (4-F) Ph analogue 15 shows
much improved ARE-LUC activity (EC50, 2.5 nM), potent AR
binding, but no androgenic activity. The movement of the F-
substituent to the 3-Ph 16 and the 2-Ph position 17 resulted in
reduced ARE-LUC agonism and AR binding activity. The same
pattern is seen for the (4-Cl) Ph 18, the (3-Cl) Ph 19, and the
(2-Cl) Ph 20 analogues, with 18 showing the most potent
ARE-LUC agonism. One should note that 18 also shows
substantial N/C-interaction assay activity (EC50, 172.2 nM), an
indicator of androgenic functional activity which one wants to
minimize with SARMs. It is important to note that binding
affinity appears to have a poor correlation with agonist potency.
For example, compounds 28 and 40 are not the most potent
binders in comparison to other analogues (Table 1), but their
agonist potencies reached into sub-nM levels. A correlation
between N/C-interaction activity and binding affinity toward
the AR LBD is even less pronounced. The disconnect between
binding and functional responses can be rationalized by the fact
that functional activity comes from cofactor protein interactions
with N- and C-termini of a full length AR in a cell while the
ligand binding itself utilizes only the LBD region of an AR
protein. We concluded that it would be more prudent to follow
SAR trends using a functional assay (ARE-LUC), which is
expected to directly translate into an in vivo outcome. The
absence of published crystallography data for ligands bound to
the full length AR protein limits researchers to interpret the
functional response observed in both ARE-LUC and N/C-
interaction assays and to correlate them with the binding pose
and affinity.
ARE-LUC
agonism
EC50 (nM)
binding
IC50
N/C
interaction
EC50 (nM)
ARE-
LUC
LipE
compd
R
(nM)
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
H
689.0
2.5
99.5
12.7
34.2
10.1
36.5
107.0
56.3
2.4
>10000
>10000
1530.4
>10000
172.2
>10000
4818.2
7.6
4.6
6.9
5.8
5.6
6.6
4.3
5.6
6.9
4.9
5.9
5.5
5.0
6.1
5.1
7.0
7.5
6.1
5.4
6.8
7.6
5.3
4.9
5.5
6.0
7.0
6.1
6.1
4-F
3-F
37.4
14.8
3.4
2-F
4-Cl
3-Cl
2-Cl
4-Br
4-CF3
275.0
6.1
0.34
1.8
33.1
145.5
133.2
179.4
16.4
60.9
14.3
9.1
>10000
624.3
>10000
>10000
71.8
4-OMe
8.6
4-COOMe
4-oxazolyl
4-CN
6.8
7.2
1.8
2-CF3
1.7
372.0
0.77
2-CN
0.15
0.074
0.40
1.1
4-F, 2-CN
4-F, 2-Br
4-F, 2-Me
4-CN, 2-Br
4-CN, 2-F
4-CN, 2-OMe
4-Cl, 2-OMe
4-OMe, 2-F
4-OMe, 2-Cl
4-Cl, 2-F
2,4-Cl2
0.74
11.4
10.7
4.8
39.5
72.2
0.060
0.039
7.6
0.41
3.2
0.14
16.1
365.8
426.7
3.5
2118.1
>10000
>10000
54.1
17.1
2.9
0.61
0.28
0.25
0.11
5.0
0.85
9.1
1.4
4-Br, 2-F
36.8
0.93
a
EC50 and IC50 values are given as the mean of at least four duplicate
experiments.
Figure 2. Docked structure of 15 (orange) in the AR ligand binding
domain. Only critical residues within the binding site are shown.
Hydrogen bonds are denoted by dotted red line.
Additional 4-substitued analogues 21−26 also show potent
ARE-LUC functional agonism with varying amounts of N/C-
interaction assay activity. The most selective compounds are 22
((4-CF3)Ph), 24 ((4-CO2Me)Ph), and 25 ((4-oxazolyl)Ph),
against Trp741 residue to form a π−π interaction and is closely
bordered by helix-12 residues Ile898 and Val903. As seen in
2464
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
Figure 3. 4-Cyanoindoline class modifications that demonstrated loss of potency. (A) changes in ARE LUC activity upon modifications of scaffold 6;
(B) inactive chemotypes.
but they show reduced LipE values compared to 15. Other
substituents and combinations of substituents showed potent
ARE-LUC agonist potency, but these compounds also showed
moderate to potent N/C-interaction activity. For example, 2-
phenyl substituted analogues 27 ((2-CF3)Ph) and 28 ((2-
CN)Ph) show potent ARE-LUC agonism but N/C-interaction
assay activity is also present. Combining optimal 4- and 2-Ph
substituents can produce compounds with very potent ARE-
LUC agonism (29 (4-F, 2-CN)Ph, 30 (4-F, 2-Br)Ph, 32 (4-
CN, 2-Br)Ph, 33 (4-CN, 2-F)Ph, 38 (4-Cl, 2-F)Ph, 39 (2,4-
Cl2)Ph, and 40 (4-Br, 2-F)Ph) but with nM or sub-nM N/C-
interaction assay activity.
Various modifications in regions of the scaffold other than
the phenyl group lead to either some loss of potency or
complete lack of binding. The (R)-1-(2-hydroxy-2-methyl-3-
phenoxypropanoyl)indoline-4-carbonitrile enantiomer shows
much weaker ARE-LUC agonist potency (Figure 3A). This is
consistent with our earlier observation where the 2-hydroxy
group of the docked R-isomer failed to make the hydrogen
bond interaction with Asn705. The introduction of a nitrogen
atom into position 5 of the indoline core (42) leads to some
loss of ARE-LUC potency while selectivity is not improved
(data not shown). Azoindoline derivatives 43−45 (Figure 3B)
showed no significant potency in ARE-LUC assay for
corresponding 4-F Ph or 2-F Ph analogues. These structural
modifications lead to changes in the electronics of the indoline
ring which have a detrimental effect on the hydrogen bond
capability of the nitrile group with Arg752 and Gln711 residues.
The replacement of the oxygen atom of the phenoxy moiety for
sulfur (46) or sulfone group also resulted in no appreciable
agonist activity for corresponding 4-F Ph or 2-F Ph analogues.
Methylation of the OH group in compound 15 led to 47, which
resulted in a complete loss of functional potency in both ARE-
LUC and N/C-interaction assays with only residual 1.2 μM
binding potency remaining. This loss of binding affinity is
consistent with the proposed docking model where the 2-
hydroxyl group makes a key interaction with Asn705 (Figure
2). The replacement of the phenyl ring in scaffold 6 with a
pyrazole group, 48, or regioisomeric pyridine analogues, 49, did
not retain the desired activity in the agonist assay. A shift of
cyano group from position 4 to position 5 of the indoline core
(50) also led to a complete loss of potency as can be expected
from a loss of a key interaction with amino acid residues
Arg752 and Gln711 in the binding pocket. The indoline core
was ring expanded to the tetrahydroisoquinoline core, 51, to
avoid possible aniline metabolites, but no appreciable activity
was found by this change. The chemistry team also attempted
to switch from the indoline core, thus removing a structural
toxicology alert associated with aniline derivatives. Tetrahy-
droisoquinoline scaffold 51 was explored as a potential
alternative for the indoline motif, however, no appreciable
potency was achieved (data not shown).
Pharmacokinetics. On the basis of its potent ARE-LUC
agonism, supra-μM N/C-interaction assay activity, and excellent
physical properties (MW, 340.3; ClogP, 2.9; TPSA, 74;
aqueous solubility at pH 6.5, 14 μM; human liver microsome
clearance, 21 μL/min/mg), 15 was selected for in vivo studies.
Compound 15 also possessed high lipophilic efficiency LipE
(Table 1) with high selectivity for ARE-LUC agonism over N/
C-interaction activity. Pharmacokinetic (PK) studies were
2465
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
conducted on 15, and the results are presented in Table 2.
Intravenous PK at 2 mg/kg showed moderate clearance (Clp,
to differences in compound distribution between the tissues
examined.
Treatment with exogenous androgens is known to repress
luteinizing hormone (LH) release from the pituitary gland
(negative feedback).18 Therefore, 15 was tested for effects on
LH in intact rats in the course of a two-week exploratory
toxicology study. An ideal SARM would exhibit limited
repression of LH release and therefore have no effect on
downstream testosterone production by the testes. In our
study, decreased LH levels were observed only at the highest
dose tested, 100 mg/kg (Figure 5), but this decrease was not
significantly different from the vehicle treated group LH levels.
Exogenous androgen administration is also known to have
effects on plasma lipids and lipoproteins.19 The effects of 3, 10,
and 100 mg/kg of 15 on these parameters in intact rats are
shown in Figure 6. The results show that 15 did not modulate
plasma levels of total cholesterol (CHOL), high-density
lipoprotein (HDL), low-density lipoprotein (LDL), or
triglycerides (TRIG) at any dose tested compared to the
vehicle treated group.
Table 2. Rat Pharmacokinetic Data for 15
2 mg/kg, IV
Clp (mL/min/kg)
Vss (L/kg)
31
6
AUC (ng·h/mL)
T1/2 (h)
1100
5
10 mg/kg, SC
Cmax (ng/mL)
Tmax (h)
166
4
AUC (ng·h/mL)
T1/2 (h)
2030
6
a
%F
37
6 mg/kg, PO
Cmax (ng/mL)
Tmax (h)
254
8
AUC (ng·h/mL)
T1/2 (h)
3750
5
b
DISCUSSION
%F
>100
■
a
AUC not completely captured due to prolonged absorption.
Bioavailability was variable ranging from 34 to 177%. Study was a
The goal of the work described in this manuscript was to
identify compounds with high ARE-LUC agonist assay potency
with minimal N/C-interaction assay potency to achieve specific
anabolic activity with minimal hypertrophic effects on the
prostate and seminal vesicles. In general, the 1-(2-hydroxy-2-
methyl-3-phenoxypropanoyl) indoline-4-carbonitrile series
showed increased N/C-interaction assay potency with increas-
ing ARE-LUC potency (Figure 7). Compound 15 was selected
for in vivo assessment, at least in part, because it had high ARE-
LUC potency for a compound with minimal N/C-interaction
assay potency.
b
parallel design and based on time−concentration profiles, AUC
differences were due to absorption differences and not clearance
differences.
31 mL/mg/kg), a moderate volume of distribution (Vss, 6 L/
kg), and a moderate half-life (T1/2, 5 h). Subcutaneous PK at 10
mg/kg showed a robust maximum concentration (Cmax, 0.48
μM), a delayed time to Cmax (Tmax, 4 h), and moderate half-life
(T1/2, 6 h). Oral PK at 6 mg/kg also showed good PK
parameters: Cmax, 0.78 μM; Tmax, 8 h, T1/2, 5 h, and variable oral
bioavailability (%F) with a mean greater than 100%.
The functional, anabolic selectivity seen with the 1-(2-
hydroxy-2-methyl-3-phenoxypropanoyl)indoline-4-carbonitrile
series was predicted experimentally through use of the N/C-
interaction assay. Steroid receptor coactivators (SRCs) and
other cofactors are known to bind to the N- and C-termini of
the androgen receptor to initiate AR signaling. Mutations in the
N- or C-terminal domains are known to result in androgen
insensitivity syndrome (AIS) and sexual developmental
abnormalities.7 Because AR N/C interdomain interactions are
required for transcriptional activation similar to that induced by
testosterone and DHT, any compounds that do not signal
through both N- and C-AR termini are expected to show
limited hypertrophic effects on the prostate and/or seminal
vesicles.
The unique biological profile of 15 can be better appreciated
by comparison with known AR agonists such as testosterone.
While compound 15 shows potent ARE-LUC activity,
testosterone demonstrated potent N/C-interaction activity
(EC50, 1.1 nM). In addition to the data presented above, the
1-(2-hydroxy-2-methyl-3-phenoxypropanoyl)indoline-4-carbon-
itrile series had no liabilities in in vitro safety assays. Potent
ARE-LUC positive compounds had IC50 > 100 μM in a hERG
dofetilide binding assay. Furthermore, the compounds showed
no effect on proliferation of a human liver cell line (THLE) at
IC50> 300 μM. Compounds also showed no significant
inhibition of cytochrome P450 enzymes (CYP1A2, CYP2C8,
CYP2C9, CYP2D6, and CYP3A4 when tested at 3 μM.
Compound 15 was tested in a two-week exploratory toxicity
study in rats and was well tolerated at doses up to 100 mg/kg/
day, the highest dose tested, for 14 consecutive days.
In Vivo Efficacy. In vivo effects of compounds were
evaluated in orchidectomized rats. The weight of prostate,
seminal vesicles, and levator ani muscle tissues was determined
after compound administration and compared with the values
for vehicle-treated animals. The effects of 3, 10, and 30 mg/kg
of 15 on the weights of the levator ani muscle, prostate, and
seminal vesicles in orchidectomized rats are shown in Figure 4.
The levator ani muscle represents a hyper-responsive skeletal
muscle in orchidectomized rats. Testosterone propionate (TP)
was included as a positive control that induces anabolic
(increased levator ani muscle weight) and androgenic
(increased prostate and seminal vesicle tissue weights)
responses in this model. After subcutaneous administration,
15 induced a statistically significant increase in the levator ani
muscle weight at all doses. Importantly, the weights of the
ventral prostate and seminal vesicles were not significantly
increased over the vehicle treated group tissue weights.
Testosterone propionate (1 mg/kg), administered subcuta-
neously, induced significant tissue weight increases in the
levator ani muscle, prostate, and seminal vesicles. This suggests
that for 15, an anabolic effect on muscle can be achieved at
doses which do not induce androgenic effects, thereby
suggesting that 15 exhibits an appropriate therapeutic index
that evaluates the compound’s anabolic versus androgenic
activity. It is important to note that the concentration of 15 in
muscle, prostate, and seminal vesicle tissues was similar (data
not shown), suggesting that the observed selectivity is not due
2466
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
Figure 4. The effects of 3, 10, and 30 mg/kg of 15 on weights of the levator ani muscle, prostate, and seminal vesicles in orchidectomized male rats
(*, p < 0.05 vs VEH). Testosterone propionate (TP).
even at the highest dose tested, 30 mg/kg,. Therefore,
compound 15 serves as a valuable tool and proof-of-concept
molecule in the development of additional SARMs with
excellent tissue selectivity. These results will be reported in
due course.
CONCLUSION
■
Selective androgen receptor modulators (SARMs) are currently
being pursued as a treatment for muscle wasting diseases with
molecules that improve upon previous clinical compounds
which showed poor tissue selectivity (i.e., androgenic effects on
the prostate and seminal vesicles), poor physical properties,
and/or toxicity. Our SARM compounds were identified using
structure-based design to derive potent anabolic agents using an
assay predictive of androgenic activity (N/C-termini interaction
assay) and by designing chemical matter in good chemical
property space. The advanced lead, compound 15, showed
excellent potency and efficacy in a predictive assay of anabolic
muscle activity (ARE-LUC EC50, 2.5 nM), no activity in an
assay of androgenic activity (N/C-interaction assay EC50,
>10000 nM), and good physicochemical properties (ClogP <
3). This compound, dosed for 4 days in an orchidectomized
immature rat model, showed anabolic activity in the levator ani
muscle and selectivity over the prostate and seminal vesicles
EXPERIMENTAL SECTION
■
Chemistry. The proton nuclear magnetic resonance (1H NMR)
spectra were recorded at 300 MHz on a Bruker DPX-300 or at 400
MHz on a Bruker DPX-400 spectrometer using tetramethylsilane (δ
0.0) as an internal standard. Mass spectra were obtained using a
Micromass platform electrospray ionization quadrapole mass spec-
trometer. Flash chromatography was performed using EM Science
230−400 mess silica gel. Thin-layer chromatography (TLC) was
performed in Analtech silica gel GHLF 250 μm prescored plates.
Analytical LC/MS was performed using an HP Agilent 1100 LC/MS
with an Agilent 1100 diode array detector (Mass spectrometer
detector: Agilent MS column, Waters Xterra MS C18 30 mm × 2.1
mm i.d., 3.5 μm. Flow rate: 1.00 mL/min. Run time: 5.00 min.
Gradient elution: 0 min 90% water, 10% acetonitrile; 3 min 10% water,
2467
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
to pH 2 with concentrated HCl, followed by the addition of NaCl till
saturation. The resulting solution was extracted with ethyl acetate (3 ×
200 mL), dried over Na2SO4, and concentrated under reduced
pressure to obtain crude mixture. This crude mixture was purified by
triturating with 5% ethyl acetate in petroleum ether to afford 9 as a
brown liquid (17.0 g, 45%). TLC system: Rf 0.1 (10% MeOH:DCM).
1H NMR (400 MHz, DMSO-d6): (major rotamer) δ 1.9 (s, 3H),
2.27−2.37 (m, 2H), 3.51−3.57 (m, 2H), 4.23−4.25 (m, 1H), 5.16 (s,
1H), 5.29 (s, 1H), 12.05 (bs, 1H); (minor rotamer): δ 1.8 (s, 3H),
2.17−2.27 (m, 2H), 3.38−3.43 (m, 2H), 4.77−4.88 (m, 1H), 5.04 (s,
1H), 5.16 (s, 1H), 12.05 (bs, 1H). LCMS (% area) 85.09; m/z = 184.2
[M + H].
(3R,8αR)-3-(Bromomethyl)-3-methyltetrahydro-1H-pyrrolo[2,1-
c][1,4]oxazine-1,4(3H)-dione 10. A solution of NBS (21.3 g, 120.0
mmol) in DMF (75 mL) was added dropwise to a solution of 9 (17.0
g, 92.8 mmol) in CCl4 (50 mL) and DMF (60 mL) in dark at 0 °C,
and stirring was continued at 0 °C for 2 h and at room temperature for
48 h. The reaction mixture was concentrated under vacuum to get a
crude mixture, and saturated aqueous NaCl solution (100 mL) was
added to this mixture and then extracted with EtOAc (3 × 200 mL).
The combined organic extracts were washed with water (6 × 200 mL),
dried over Na2SO4, and concentrated under reduced pressure to
obtain 10 as a white solid (7.0 g, 29%). TLC system: Rf 0.6 (10%
Figure 5. Effect of 3, 10, and 100 mg/kg of 15 on serum luteinizing
hormone concentration in intact rats administered with 15 for 14
consecutive days.
90% acetonitrile. Column temperature: 50.0 °C. UV signals: 215 nm,
254 nm. MS parameters: mass range 100−1000, fragmentor 140, Gain
EMV 1.0.). Preparative RP-HPLC was performed using a Gilson
HPLC (Phenomenex LUNA C18 column: 60 mm × 21.20 mm i.d., 5
μm particle size, and eluted with an acetonitrile/water gradient
(containing either 0.2% Et3N or 0.1% HCO2H)). All final compounds
were >95% pure by HPLC.
The synthesis of compound 11 was done according to a modified
procedure described in Kirkovsky et al.20 The synthetic protocols to
prepare chemotypes described in Figure 3 are disclosed in Supporting
Information.21
1
MeOH in DCM). H NMR (300 MHz, DMSO-d6): δ 1.57 (s, 3H),
1.77−2.05 (m, 3H), 2.22−2.30 (m, 1H), 3.31−3.43 (m, 1H), 3.46−
3.56 (m, 1H), 3.87, 4.02 (ABq, 2H, J = 11.1 Hz), 4.70 (dd, 1H, J = 6.9,
9.6 Hz). LCMS (% area) 96.53; m/z = 264.0 [M + 2].
(R)-3-Bromo-2-hydroxy-2-methylpropanoic Acid 11. A solution of
10 (7.0 g, 26.70 mmol) in 24% aqueous HBr (120 mL) was refluxed
for 2 h. After cooling to room temperature, solid NaCl was added until
saturation, and the mixture was extracted with ethyl acetate (4 × 150
mL). The combined ethyl acetate layer was extracted with saturated
aqueous NaHCO3 solution (4 × 150 mL). The aqueous layer was
acidified to pH 1 with concentrated HCl and then extracted with
EtOAc (4 × 150 mL). The combined EtOAc extracts were dried over
Na2SO4 and concentrated under reduced pressure to obtain 11 as a
white solid (2.3 g, 48%). The enantiomeric purity of compound 11
(derived from D-proline) was established by comparing compound 15
on chiral HPLC to its stereoisomer 41 that was synthesized starting
from L-proline (see Supporting Information). TLC system: Rf 0.2
(R)-1-Methacryloylpyrrolidine-2-carboxylic Acid 9. A solution of
methacryloyl chloride (17.3 mL, 177 mmol) in acetone (100 mL) was
added to a solution of D-proline (20.0 g, 173.7 mmol) in 2 M aqueous
NaOH (100 mL) and acetone (100 mL) dropwise at 0 °C. The pH of
the reaction mixture was maintained within 10.3
0.3 range
(monitored by pH meter) via simultaneous dropwise addition of 2
M aqueous NaOH during the addition of methacryloyl chloride. The
reaction mixture was allowed to warm up to room temperature and
stirred for 6 h at room temperature. Acetone was removed from the
reaction mixture under vacuum, and the resulting mixture was acidified
1
(10% MeOH in DCM). H NMR (300 MHz, DMSO-d6): δ 1.36 (s,
3H), 3.53, 3.64 (ABq, 2H, J = 10.2 Hz), 5.5 (bs, 1H), 12.9 (bs, 1H).
LCMS (% area) 96.53; m/z = 181.19 [M − 1].
Figure 6. The effect of 3, 10, and 100 mg/kg of 15 on serum concentrations of total cholesterol (CHOL), high-density lipoprotein (HDL), low-
density lipoprotein (LDL), and triglycerides (TRIG) in intact rats administered with 15 for 14 consecutive days.
2468
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
Figure 7. pARE-LUC EC50 (M) vs pN/C interaction EC50 (M).
Figure 8. In vitro profile of lead compound 15 compared to testosterone.
(R)-1-(3-Bromo-2-hydroxy-2-methylpropanoyl)indoline-4-car-
bonitrile 12. Thionyl chloride (0.65 mL, 9.0 mmol) was added
dropwise to a solution of 11 (1.4 g, 7.6 mmol) in dry
dimethylacetamide (8 mL) at −12 °C under nitrogen atmosphere,
and the stirring was continued for 3 h at same temperature. A solution
of indoline-4-carbonitrile (1.0 g, 6.9 mmol; see Supporting
Information) in dry dimethylacetamide (8 mL) was added to this
reaction mixture. The reaction mixture was allowed to warm up to
room temperature, and stirring was continued for 16 h. The reaction
mixture was diluted with ethyl acetate (150 mL) and washed with
water (2 × 100 mL). The organic layer was dried over Na2SO4 and
concentrated under reduced pressure to obtain 12 as an off-white solid
(1.4 g, 66%). TLC system: Rf 0.35 (30% ethyl acetate in petroleum
DMSO-d6): δ 1.56 (s, 3H), 2.92 (d, 1H, J = 5.7 Hz), 3.07 (d, 1H, J =
4.8 Hz), 3.29−3.36 (m, 2H), 4.26−4.33 (m, 2H), 7.39 (t, 1H, J = 7.5),
7.5 (d, 1H, J = 7.8 Hz), 8.23 (bd, 1H, J = 7.2 Hz). LCMS (% area)
96.27; m/z = 229.1 [M + H]. Chiral HPLC: 99.76% (RT 11.79 min.).
1-(3-(4-Fluorophenoxy)-2-hydroxy-2-methylpropanoyl)indoline-
4-carbonitrile 15. K2CO3 (0.84 g, 6.1 mmol) was added to a mixture
of 13 (0.70 g, 3.1 mmol) and 4-fluorophenol (0.45 g, 4.0 mmol) in 2-
butanone (12 mL), and the reaction mixture was refluxed for 48 h.
The reaction mixture was diluted with ethyl acetate, washed with 1N
NaOH, water, dried over Na2SO4, and concentrated under reduced
pressure to obtain crude mixture. This crude mixture was purified by
triturating with pentane to afford 15 as a pale-yellow solid (0.70 g,
1
67%). TLC system: Rf 0.3 (DCM). H NMR (300 MHz, DMSO-d6):
1
ether). H NMR (400 MHz, DMSO-d6): δ 1.51 (s, 3H), 3.26−3.31
δ 1.48 (s, 3H), 3.27−3.32 (m, 2H), 3.99 (d, 1H, J = 9.3 Hz), 4.23 (d,
1H, J = 9.9 Hz), 4.46−4.50 (m, 1H), 4.66−4.73 (m, 1H), 5.92 (s, 1H),
6.95−6.99 (m, 2H), 7.06−7.13 (m, 2H), 7.35 (t, 1H, J = 7.5 Hz), 7.45
(d, 1H, J = 6.9 Hz), 8.32 (d, 1H, J = 8.4 Hz). LCMS (% area) 95.28;
m/z = 341.1 [M + H]. Chiral HPLC: 99.75% (RT 11.98 min.).
Biology. Materials. Intact or castrated male Sprague−Dawley rats
were purchased from Charles River Laboratories (Wilmington, MA).
Testosterone propionate and DHT were purchased from Sigma-
Aldrich (St. Louis, MO), [3H]-mibolerone from PerkinElmer,
Lipofectamine 2000 transfection reagent from Invitrogen (Carlsbad,
CA), phenol red free DMEM/high glucose, nonessential amino acids,
(m, 2H), 3.66, 3.89 (ABq, 2H, J = 10.4 Hz), 4.45−4.49 (m, 1H),
4.54−4.61 (m, 1H), 6.02 (s, 1H), 7.36 (t, 1H, J = 8 Hz), 7.46 (d, 1H,
7.6 Hz), 8.34 (d, 1H, J = 8 Hz). LCMS (% area) 88.19; m/z 310.8 [M
+ H].
(S)-1-(2-Methyloxirane-2-carbonyl)indoline-4-carbonitrile 13.
K2CO3 (1.9 g, 13.5 mmol) was added to a solution of 12 (1.4 g, 4.5
mmol) in 2-butanone (15 mL), and the reaction mixture was stirred
for 6 h. The reaction mixture was filtered, and the filtrate was
concentrated to give 13 as a pale-yellow solid (0.82 g, 80%). TLC
system: Rf 0.3 (30% EtOAc in petroleum ether). 1H NMR (300 MHz,
2469
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
penicillin−streptomycin from Gibco (Grand Island NY), and fetal
bovine serum (FBS) from Atlanta Biologicals (Norcross, GA). CV-1
and COS cells were obtained from American Tissue Culture
Collection (Manassas, VA).
once daily. Intact (sham) rats served as controls and were not treated.
On the day of the last administration, the rats were euthanized by CO2
asphyxiation and blood was collected by cardiac puncture. Prostate,
seminal vesicles, and levator ani muscle tissues were collected and
weighed. Circulating LH concentrations in plasma were determined
with an LH ELISA kit from Endocrine Technologies, Newark, CA.
The concentration of test compounds in plasma, the levator ani
muscle, and prostate were determined in tissue samples solubilized in
water (4× volume per gram of tissue).
Cell-Based Assays. The binding and reporter assays were performed
on cells transfected in T225 cm2 flasks with the plasmids described
below, harvested, frozen, and replated in 96-well plates. For the AR
binding assay, COS cells were transfected with phAR, an expression
vector coding for full length human AR cDNA in pcDNA3
(Invitrogen). For the ARE-LUC reporter assay, CV-1 cells were
transfected with: (1) phAR and (2) pARE-Luciferase, a luciferase
reporter plasmid containing an androgen response element (ARE)
DNA sequence upstream of a firefly luciferase reporter in pGL3-Basic
vector (Promega, Madison, WI). For the 2-hybrid AR N/C-interaction
reporter assay CV-1 cells were transfected with three plasmids: (1)
pVP16-hAR-NTD, expressing human AR amino acids 1−561 in frame
to produce a fusion protein with VP16 in pVP16 vector (Clontech,
Mountain View, CA), (2) pGAL4-hAR-LBD, a plasmid containing the
human AR-LBD, AR amino acids 644−919, fused in frame with the
GAL4 DNA binding domain within the p133 vector (Promega), (3)
p5XGAL4-Luciferase reporter plasmid containing five copies of the
GAL4 upstream activation sequence followed by a minimal E1b
promoter controlling firefly luciferase expression in pGL3-Basic. Cells
were maintained in phenol red DMEM/high glucose, 10% non heat-
inactivated fetal bovine serum, 1% nonessential amino acids, 1%
penicillin−streptomycin at 37 °C in 5% CO2 and a humidified
atmosphere. For transfections, cells were plated at 11−12 million cells
per T225 cm2 flask one or two days before the transfection procedure.
Cells were transfected using Lipofectamine-2000 at a plasmid DNA to
Lipofectamine ratio of 1:3 (μg:μL) in phenol red free DMEM/high
glucose, 1% sodium pyruvate, 1% nonessential amino acids, and 1%
GlutaMAX-I (Gibco). After a 4 h incubation, cells were trypsinized,
frozen, and stored in cryovials at −150 °C.
For the luciferase reporter assays transfected, frozen CV-1 cells were
plated at 40000 cells per well in 96-well plates and incubated for 6 h.
Testosterone, DHT, and test compounds were resuspended and
serially diluted in DMSO and added to cells in phenol red free
DMEM/high glucose, 1% sodium pyruvate, 1% nonessential amino
acids, 1% GlutaMAX-I, 5% charcoal stripped fetal bovine serum, 1%
penicillin−streptomycin. After 17 h at 37 °C and 5% CO2, cells were
lysed in lysis reagent (Promega). Lysate samples were analyzed for
luciferase activity in Microlite microtiter plates (Thermo Labsystems)
using luciferase reagent (Promega) and a Victor 2 1420 multilabel
counter luminometer (Perkin-Elmer, Waltham, MA). All reporter
assays were performed in duplicate wells at least three times.
ASSOCIATED CONTENT
* Supporting Information
■
S
Supporting experimental details and characterization for
reported compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: 617-665-5621. Fax: 860-686-5499. E-mail: eugene.
chekler@pfizer.com.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge Dr. K. Vishwanathan for conducting the in
vivo experiments. We would also like to thank Dr. B. Hu and
Dr. J. Wrobel for early synthetic contributions. We thank Dr. A.
Villalobos for supporting the project.
ABBREVIATIONS USED
■
HATU, 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo-
[4,5-b]pyridinium 3-oxid hexafluorophosphate; DIPEA, diiso-
propylethylamine; DHT, dihydrotestosterone; TPSA, total
polar surface area
REFERENCES
■
(1) Lu, N. Z.; Wardell, S. E.; Burnstein, K. L.; Defranco, D.; Fuller, P.
J.; Giguere, V.; Hochberg, R. B.; McKay, L.; Renoir, J.-M.; Weigel, N.
L.; Wilson, E. M.; McDonnell, D. P.; Cidlowski, J. A. International
Union of pharmacology. LXV. The pharmacology and classification of
the nuclear receptor superfamily: glucocorticoid, mineralocorticoid,
progesterone, and androgen receptors. Pharm. Rev. 2006, 58, 782−
797.
Binding Assay. Full length AR transfected COS cells were thawed
and plated in 96-well isoplates (PerkinElmer) at 50000 cells per well
and incubated for 18 h at 37 °C, 5% CO2 in phenol red free DMEM/
high glucose containing 1% sodim pyruvate, 1% nonessential amino
acids, 1% GlutaMAX-I, 5% charcoal stripped FBS, and 1% penicillin−
streptomycin. Testosterone, DHT, and test compounds were
suspended and serially diluted in DMSO. Cells were treated with
dilutions of each compound plus 0.5 nM [3H]-mibolerone and
incubated for 3.5 h at 37 °C, 5% CO2. Cells were then washed 3 times
with culture media and lysed with 25 μL of 0.5% SDS. Optiphase
Supermix scintillation cocktail (Perkin-Elmer) was added, 0.1 mL per
well, and radioactivity was measured as disintegrations per minute
(DPM) on a Wallac 1450 Microbeta liquid scintillation and
luminescence counter. Dose−response curves and IC50 values were
generated using XLfit software.
(2) Kilbourne, E. J.; Moore, W. J.; Freedman, L. P.; Nagpal, S.
Selective androgen receptor modulators for frailty and osteoporosis.
Curr. Opin. Invest. Drugs 2007, 8, 821−829.
(3) Gao, W.; Bohl, C. E.; Dalton, J. T. Chemistry and structural
biology of androgen receptor. Chem. Rev. 2005, 105, 3352−3370.
(4) (a) Bhasin, S.; Jasuja, R. Selective androgen receptor modulators
as function promoting therapies. Curr. Opin. Clin. Nutr. Metab. Care
2009, 12, 232−240. (b) Kilbourne, E. J.; Kenney, T.; Chippari, S.;
McNally, C.; Wang, Y.; Lam, H.-S.; Vishwanathan, K.; Nagpal, S.;
Thompson, C. C.; Chekler Piatnitski, E. L. Probing androgen receptor
co-factor selectivity profiles:a chemical tool to determine cross-talk
between androgen receptor and β-catenin in vivo. Med. Chem.
Commun. 2013, 4 (3), 582−589.
(5) Dubois, V.; Laurent, M.; Boonen, S.; Vanderschueren, D.;
Claessens, F. Androgens and skeletal muscle: cellular and molecular
action mechanisms underlying the anabolic actions. Cell. Mol. Life Sci.
2012, 69, 1651−1667.
(6) Mohler, M. L.; Bohl, C. E.; Narayanan, R.; He, Y.; Hwang, D. J.;
Dalton, J. T.; Miller, D. D. Nonsteroidal tissue-selective androgen
receptor modulators. Methods Principals Med. Chem. 2008, 39, 249−
304.
In Vivo Studies. All procedures conducted in this study were done
in accordance with the Pfizer Animal Care and Use Committee and
the National Institutes of Health Guide for the Care and Use of
Laboratory Animals. All animals were provided ad libitum with water
and standard rodent chow containing 28.5% protein, 13.5% fat, and
58% carbohydrate as a percentage of calories (Purina, LabDiet).
Orchidectomized Sprague−Dawley male rats were grouped six rats per
group and dosed for 4 days (∼7 days after surgery), with vehicle, test
compounds at 3, 10, or 30 mg/kg in vehicle, or 1 mg/kg testosterone
propionate in 10% ethanol in corn oil administered subcutaneously
(7) Ghali, S. A.; Gottlieb, B.; Lumbroso, R.; Beitel, L. K.; Elhaji, Y.;
Wu, J.; Pinsky, L.; Trifiro, M. A. The use of androgen receptor amino/
2470
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471

Journal of Medicinal Chemistry
Article
carboxyl-terminal interaction assays to investigate androgen receptor
gene mutations in subjects with varying degrees of androgen
insensitivity. J. Clin. Endocrin. Metab. 2003, 88, 2185−2193.
(8) Schmidt, A.; Harada, S.-I.; Kimmel, D. B.; Bai, C.; Chen, F.;
Rutledge, S. J.; Vogel, R. L.; Scafonas, A.; Gentile, M. A.; Nantermet, P.
V.; McElwee-Witmer, S.; Pennypacker, B.; Masarachia, P.; Sahoo, S.
P.; Kim, Y.; Meissner, R. S.; Hartman, G. D.; Duggan, M. E.; Rodan, G.
A.; Towler, D. A.; Ray, W. J. Identification of Anabolic Selective
Androgen Receptor Modulators with Reduced Activities in Repro-
ductive Tissues and Sebaceous Glands. J. Biol. Chem. 2009, 284,
36367−36376.
Bicalutamide Analogues Bearing Electrophilic Groups in the B
Aromatic Ring. J. Med. Chem. 2000, 43 (4), 581−590.
(21) Chekler Piatnitski, E. L.; Khan, T. A.; Anderson, J. T.; Tangirala,
R.; Verhoest, P. R.; Gilbert, A. M. Azaindolines: derisking the indoline
structural alert. Tetrahedron Lett. 2012, 53, 377−379.
(9) Jones, A.; Hwang, D. J.; Duke, C. B., III; He, Y.; Siddam, A.;
Miller, D. D.; Dalton, J. T. Nonsteroidal selective androgen receptor
modulators enhance female sexual motivation. J. Pharm. Exp. Ther.
2010, 334, 439−448.
(10) Miller, C. P.; Shomali, M.; Lyttle, C. R.; O’Dea, L. S. L.;
Herendeen, H.; Gallacher, K.; Paquin, D.; Compton, D. R.; Sahoo, B.;
Kerrigan, S. A.; Burge, M. S.; Nickels, M.; Green, J. L.;
Katzenellenbogen, J. A.; Tchesnokov, A.; Hattersley, G. Design,
synthesis, and preclinical characterization of the selective androgen
receptor modulator (SARM) RAD140. J. Med. Chem. Lett. 2011, 2,
124−129.
(11) Nirschl, A. A.; Zou, Y.; Krystek, S. R.; Sutton, J. C.; Simpkins, L.
M.; Lupisella, J. A.; Kuhns, J. E.; Seethala, R.; Golla, R.; Sleph, P. G.;
Beehler, B. C.; Grover, G. J.; Egan, D.; Fura, A.; Vyas, V. P.; Li, Y.-X.;
Sack, J. S.; Kish, K. F.; An, Y.; Bryson, J. A.; Gougoutas, J. Z.; Di, M. J.;
Zahler, R.; Ostrowski, J.; Hamann, L. G. N-Aryl-oxazolidin-2-imine
muscle selective androgen receptor modulators enhance potency
through pharmacophore reorientation. J. Med. Chem. 2009, 52, 2794−
2798.
(12) Vajda, E. G.; Lopez, F. J.; Rix, P.; Hill, R.; Chen, Y.; Lee, K.-J.;
O’Brien, Z.; Chang, W. Y.; Meglasson, M. D.; Lee, Y.-H.
Pharmacokinetics and pharmacodynamics of LGD-3303 [9-chloro-2-
ethyl-1-methyl-3-(2,2,2-trifluoroethyl)-3H-pyrrolo[3,2-f ]quinolin-
7(6H)-one], an orally available nonsteroidal-selective androgen
receptor modulator. J. Pharm. Exp. Ther. 2009, 328, 663−670.
(13) Nique, F.; Hebbe, S.; Peixoto, C.; Annoot, D.; Lefrancois, J.-M.;
Duval, E.; Michoux, L.; Triballeau, N.; Lemoullec, J.-M.; Mollat, P.;
Thauvin, M.; Prange, T.; Minet, D.; Clement-Lacroix, P.; Robin-
Jagerschmidt, C.; Fleury, D.; Guedin, D.; Deprez, P. Discovery of
diarylhydantoins as new selective androgen receptor modulators. J.
Med. Chem. 2012, 55, 8225−8235.
(14) Nique, F.; Hebbe, S.; Triballeau, N.; Peixoto, C.; Lefrancois, J.-
M.; Jary, H.; Alvey, L.; Manioc, M.; Housseman, C.; Klaassen, H.; Van,
B. K.; Guedin, D.; Namour, F.; Minet, D.; Van, d. A. E.; Feyen, J.;
Fletcher, S.; Blanque, R.; Robin-Jagerschmidt, C.; Deprez, P.
Identification of a 4-(hydroxymethyl)diarylhydantoin as a selective
androgen receptor modulator. J. Med. Chem. 2012, 55, 8236−8247.
(15) Kirkovsky, L.; Mukherjee, A.; Yin, D.; Dalton, J. T.; Miller, D. D.
Chiral Nonsteroidal Affinity Ligands for the Androgen Receptor. 1.
Bicalutamide Analogues Bearing Electrophilic Groups in the B
Aromatic Ring. J. Med. Chem. 2000, 43, 581−590.
(16) Maestro; Schrodinger, LLC: Portland, OR; http://www.
̈
schrodinger.com/.
(17) Bohl, C. E.; Miller, D. D.; Chen, J.; Bell, C. E.; Dalton, J. T.
Structural basis for accommodation of nonsteroidal ligands in the
androgen receptor. J. Biol. Chem. 2005, 280 (45), 37747−37754.
(18) Matsumoto, A. M. Effects of chronic testosterone administration
in normal men: safety and efficacy of high dosage testosterone and
parallel dose-dependent suppression of luteinizing hormone, follicle-
stimulating hormone, and sperm production. J. Clin. Endocrinol. Metab.
1990, 70, 282−7.
(19) Zgliczynski, S.; Ossowski, M.; Slowinska-Srzednicka, J.;
Brzezinska, A.; Zgliczynski, W.; Soszynski, P.; Chotkowska, E.;
Srzednicki, M.; Sadowski, Z. Effect of testosterone replacement
therapy on lipids and lipoproteins in hypogonadal and elderly men.
Atherosclerosis 1996, 121, 35−43.
(20) Kirkovsky, L.; Mukherjee, A.; Yin, D.; Dalton, J. T.; Miller, D. D.
Chiral Nonsteroidal Affinity Ligands for the Androgen Receptor. 1.
2471
dx.doi.org/10.1021/jm401625b | J. Med. Chem. 2014, 57, 2462−2471